Streaming: what does the evidence say?

Streaming, it seems, is making a comeback in our schools. According to Department for Education research, more schools are turning to this form of attainment grouping in response to learning gaps created by Covid.

The report, published in January, found some limited evidence that streaming had been brought in by a number of schools in order to support lower-attaining pupils in the wake of the pandemic - whether staff would normally agree with the practice or not.

Indeed, the report quotes one secondary teacher as saying: “We streamed all of our year groups, which is not something I believe in, but I think that in this instance it was the right thing to do.”

The idea was to allow teachers to offer specific support for those pupils who were deemed to need it - resulting in slightly larger classes at the “higher academic end” and smaller classes at the “weaker” end.

But what exactly is streaming, and how does it differ from setting?

According to Ed Baines, a senior lecturer in psychology of education at the UCL Institute of Education, there is a fair bit of confusion around the definition. 

Streaming and setting: what is the difference?

“Typically, streaming is grouping students on the basis of a perception of their ‘ability’ across all subjects, and with a view to forming classes of students that are around the same broad level of ability - and where this grouping has all of their lessons together across all or most curriculum subjects,” he explains. 

This is different to setting, which places students into attainment groupings for specific subjects. 

Streaming is a controversial approach, says Baines, because it can overlook learning potential. 

“Any two students might score at similar levels on a test but one might have greater potential to develop than the other - and it also makes assumptions about notions of general intelligence, which are problematic and not supported by a research literature,” he says. 
 
Across education, there are many different interpretations of streaming, which can be selectively implemented. For example, a school might have streaming for the high-attaining students and mixed-attainment groups for everyone else - or have just a low-attaining group and so on.
 
“It can get very difficult to establish an evidence base relative to all the varied approaches,” says Baines. 
 
So what do we know? Is there any evidence to suggest that widespread streaming within a school can bolster attainment - particularly for students who are struggling academically?

Read more:

• Four questions you need to ask about ability grouping
• Three tips for teaching classes with a range of abilities
• Tes focus on…Closing the attainment gap

Becky Taylor is a principal research fellow at the UCL Institute of Education, an expert in setting and streaming, and co-author of the Dos and don’ts of attainment grouping guide. She warns that, actually, research has found the opposite: that streaming can limit progress. 
 
“Some teachers argue that grouping pupils with others of similar ‘ability’ means that they can target teaching more specifically to support or to challenge,” she says. “But over many years, research has consistently shown that pupils from disadvantaged backgrounds…are more likely to be allocated to lower sets and streams.”
 
Researchers, including Taylor, have also found evidence that certain minority ethnic groups and girls are more likely to be misallocated to lower ability groups than their previous attainment would have predicted. 
 
“We had a similar finding for English, but in that case, it was boys who were placed in lower groups,” says Taylor. “Another disadvantage is that lower groups tend to be assigned less experienced or less highly qualified teachers, and don’t get access to the same rich curriculum as taught to pupils in higher groups. This results in pupils in lower-attaining groups making less progress than pupils in higher sets and streams.”

The pitfalls of streaming

Another big pitfall of streaming is that it “isn’t very specific”, she continues. 

“Streaming is based on the assumption that pupils will be working at a similar level across all subjects, but this may not be the case,” she says. “Setting, where pupils are grouped on a subject by subject basis, is more flexible, but there can still be problems when grouping takes into account factors other than prior attainment, because that’s when bias can creep in.”
 
There is also research to suggest that, once pupils are in a particular group, it can be very difficult for them to move out of that group - even if they’ve made progress. 
 
“That’s the same for setting and streaming,” Taylor says. “The research suggests that teachers tend to overestimate the amount of movement between groups and we found that in some schools the structure of groups and timetables can make it almost impossible.”
 
Rob Webster, director of education research, innovation and consultancy at the University of Portsmouth’s School of Education, agrees that, in practice, schools often fail to build flexibility into their streaming approach. 
 
“If you are going to be streaming on the basis of attainment then you have to make sure you are genuinely doing that - and that includes making sure that you are allowing kids to move up and down [through the groups],” he says.
 
Another concern with this rigidity is that the attainment group into which a child is streamed will then often become their social group, since these are the children with which they are spending most of their time.
 
“Therefore, if you’re in the bottom set, then you’re socialised with that group too, not just inside the classroom, but outside as well,” he says. “So you end up with a sort of unintended, but nevertheless quite palpable social segregation. Kids in the high streams are learning and socialising with each other, and so are the kids in the bottom set.”
 
There are a number of reasons why this can be problematic, Webster says, particularly at primary level, where children are in earlier developmental stages. 
 
“One example I always fall back on is speech and language,” he explains. “You will have kids who have speech and language difficulties learning and socialising with each other, so they might not have any good models for good speech and language because of that.”
 
Schools that are considering going down the streaming route need to think about improving the social mix of different streams, he stresses. 

“Yes, we need to look at the porosity between groups on the basis of attainment, but we also need to be mindful of unintentionally setting up different environments and different worlds where kids end up being less exposed to different social groups,” he says. 
 
Another worry is that, in addition to not benefitting all pupils academically, streaming can lead to stigma, particularly for those who are streamed into lower ability groups. 
 
“We found evidence of Year 7 pupils feeling really hopeless about being stuck in low groups and unable to move up, even when they were working really hard,” says Taylor. “Some of them were ready to give up and not try at all.”

Should schools avoid streaming?

What’s more, research shows that streaming can have a negative effect on pupils with SEND, too. 

Peter Blatchford, an emeritus professor of psychology and education, is the co-author of a 2018 Institute of Education study, which illustrated that as a result of setting or streaming, SEND pupils spent a lot of time with teaching assistants and became separated from their teachers. 
  
“In mainstream secondary schools, pupils with SEND were almost always assigned to low-attaining sets in all subjects that are set and we worried about how, despite being in an ostensibly comprehensive school system, they were effectively experiencing a kind of streaming,” he explains.

For Blatchford, the “general problem” with streaming is that “it’s predicated on a notion of fixed abilities or intelligence, which is largely discredited now, by psychology at least”. 
 
“Development is far too fluid and changeable to fix at age 11, and there is a body of psychology directed at ways in which pupils, once assigned to [lower-ability streams], can become stuck and fall in line with how teachers seem to perceive their abilities,” he says. “This confirms teachers’ original selection, and is at the heart of the self-fulfilling prophecy.”

Does all of this mean that schools should be avoiding streaming at all costs, then? 

While there are clearly issues with streaming, grouping by attainment remains widespread - particularly in secondary schools - and Covid recovery presents a unique set of circumstances. Even schools that had not previously used streaming may be considering implementing it now. 

So, how can those schools minimise the risks?

According to the Education Endowment Foundation, it is “vital” that schools that do opt to implement streaming consider how the approach will enable more effective teaching for all pupils, including those with lower attainment levels. 

“It is important to ensure that all pupils follow a challenging curriculum, including lower attaining pupils,” the EEF says, adding that “ensuring flexibility in grouping arrangements and regular monitoring of learning will minimise the risk of misallocation for pupils that learn at different rates”.